Recherche avancée

Sur d’autres sites (4753)

• Convert image sequence to video using ffmpeg and list of files

  13 mai 2015, par rensa

  I have a camera taking time-lapse shots every 2–3 seconds, and I keep a rolling record of a few days’ worth. Because that’s a lot of files, I keep them in subdirectories by day and hour :

  images/
  
      2015-05-02/
  
          00/
  
              2015-05-02-0000-02
  
              2015-05-02-0000-05
  
              2015-05-02-0000-07
  
          01/
  
              (etc.)
  
      2015-05-03/

  I’m writing a script to automatically upload a timelapse of the sunrise to YouTube each day. I can get the sunrise time from the web in advance, then go back after the sunrise and get a list of the files that were taken in that period using find :

  touch -d "$SUNRISE_START" sunrise-start.txt
  
  touch -d "$SUNRISE_END" sunrise-end.txt
  
  find images/"$TODAY" -type f -anewer sunrise-start.txt ! -anewer sunrise-end.txt

  Now I want to convert those files to a video with ffmpeg. Ideally I’d like to do this without making a copy of all the files (because we’re talking 3.5 GB per hour of images), and I’d prefer not to rename them to something like image000n.jpg because other users may want to access the images. Copying the images is my fallback.

  But I’m getting stuck sending the results of find to ffmpeg. I understand that ffmpeg can expand wildcards internally, but I’m not sure that this is going to work where the files aren’t all in one directory. I also see a few people using find’s --exec option with ffmpeg to do batch conversions, but I’m not sure if this is going to work with image sequence input (as opposed to, say, converting 1000 images into 1000 single-frame videos).

  Any ideas on how I can connect the two—or, failing that, a better way to get files in a date range across several subdirectories into ffmpeg as an image sequence ?

• How to adjust mpeg 2 ts start time with ffmpeg ?

  29 juin 2015, par Maxim Kornienko

  I’m writing simple HLS (Http Live Streaming) java server to live cast (really live, not on demand) screenshow + voice. I constantly get chunks of image frames and audio samples as input to my service and produce mpeg 2 ts files + m3u8 playlist web page as output. The workflow is the following :

  1. Collect (buffer) source video frames and audio for certain period of time
  2. Convert series of video frames to h.264 encoded video file
  3. Convert audio samples to mp3 audio file

  4. Merge them to .ts file with ffmpeg command

     ffmpeg -i audio.mp3 -i video.mp4 -f mpegts -c:a copy -c:v copy -vprofile main -level:v 4.0 -vbsf h264_mp4toannexb -flags -global_header segment.ts

  5. Publish several .ts files on m3u8 playlist.

  The problem is resulting playlist interrupts after first segment is played. VLC logs following error :

  freetype error: Breaking unbreakable line
  
  ts error: libdvbpsi (PSI decoder): TS discontinuity (received 0, expected 4) for PID 17
  
  ts error: libdvbpsi (PSI decoder): TS duplicate (received 0, expected 1) for PID 0
  
  ts error: libdvbpsi (PSI decoder): TS duplicate (received 0, expected 1) for PID 4096
  
  core error: ES_OUT_SET_(GROUP_)PCR is called too late (pts_delay increased to 1000 ms)
  
  core error: ES_OUT_RESET_PCR called
  
  core error: Could not convert timestamp 185529572000
  
  ts error: libdvbpsi (PSI decoder): TS discontinuity (received 0, expected 4) for PID 17
  
  ts error: libdvbpsi (PSI decoder): TS duplicate (received 0, expected 1) for PID 0
  
  ts error: libdvbpsi (PSI decoder): TS duplicate (received 0, expected 1) for PID 4096
  
  core error: ES_OUT_SET_(GROUP_)PCR is called too late (jitter of 8653 ms ignored)
  
  core error: Could not get display date for timestamp 0
  
  core error: Could not convert timestamp 185538017000
  
  core error: Could not convert timestamp 185538267000
  
  core error: Could not convert timestamp 185539295977
  
  ...

  I guess the reason is that start time of segments do not belong to one stream, but it’s impossible to concat and resegment (with ffmepg -f segment) whole stream once new chunk is added. Tried adding #EXT-X-DISCONTINUITY tag to playlist as suggested here but it didn’t help. When I ffprobe them I get :

  Input #0, mpegts, from '26.ts':
  
  Duration: 00:00:10.02, start: 1.876978, bitrate: 105 kb/s
  
  Program 1
  
  Metadata:
  
    service_name    : Service01
  
    service_provider: FFmpeg
  
  Stream #0:0[0x100]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p, 640x640, 4 fps, 4 tbr, 90k tbn, 8 tbc
  
  Stream #0:1[0x101]: Audio: mp3 ([3][0][0][0] / 0x0003), 48000 Hz, mono, s16p, 64 kb/s   

  Where start value in line Duration: 00:00:10.02, start: 1.876978, bitrate: 105 kb/s is more or less equal for all segments.
  When I check segments from available proven-to-work playlists (like http://vevoplaylist-live.hls.adaptive.level3.net/vevo/ch1/appleman.m3u8) they all have diffrenet start values for each segment, for example :

  Input #0, mpegts, from 'segm150518140104572-424570.ts':
  
  Duration: 00:00:06.17, start: 65884.808689, bitrate: 479 kb/s
  
  Program 257
  
  Stream #0:0[0x20]: Video: h264 (Constrained Baseline) ([27][0][0][0] / 0x001B), yuv420p, 320x180 [SAR 1:1 DAR 16:9], 30 fps, 29.97 tbr, 90k tbn, 60 tbc
  
  Stream #0:1[0x21]: Audio: aac (LC) ([15][0][0][0] / 0x000F), 44100 Hz, stereo, fltp, 115 kb/s
  
  Stream #0:2[0x22]: Data: timed_id3 (ID3  / 0x20334449)

  and the next after it

  Input #0, mpegts, from 'segm150518140104572-424571.ts':
  
  Duration: 00:00:06.22, start: 65890.814689, bitrate: 468 kb/s
  
  Program 257
  
  Stream #0:0[0x20]: Video: h264 (Constrained Baseline) ([27][0][0][0] / 0x001B), yuv420p, 320x180 [SAR 1:1 DAR 16:9], 30 fps, 29.97 tbr, 90k tbn, 60 tbc
  
  Stream #0:1[0x21]: Audio: aac (LC) ([15][0][0][0] / 0x000F), 44100 Hz, stereo, fltp, 124 kb/s
  
  Stream #0:2[0x22]: Data: timed_id3 (ID3  / 0x20334449)

  differ in the way that start time of segm150518140104572-424571.ts is equal to start time + duration of segm150518140104572-424570.ts.

  How could this start value be adjusted with ffmpeg ? Or maybe my whole aproach is wrong ? Unfortunately I couldn’t find on the internet working example of live (not on demand) video service implemented with ffmepg.

• Is celery a good choice for long running tasks ?

  2 décembre 2024, par jasoos

  We are using a combination of django, rabbitmq, celery and ffmpeg to read camera streams and break them into images to be stored into filesystem. This setup works 24x7. Now, for each camera stream we are creating a separate task and each will theoretically run for indefinite period.

  


  If a stream goes down, we wait for n number of frames, create an exception and in the exception handler, after creating a delay of 1 min using time.sleep we rerun the ffmpeg process.

  


  My questions are,
Is this a right approach ?
Should we use celery for reading streams ?
Is celery the right tool to use for this task ?
Can we create delay in celery task using time.sleep ? Will it affect the other tasks ?